Back to search results
Llama-3.1-8B-Instruct · Q5_K_S
NVIDIA RTX 509032GBOllama v0.23L2 Cross-framework verified
This page aggregates 1 real-world runs of Llama-3.1-8B-Instruct (Q5_K_S) on NVIDIA RTX 5090 with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.
220.5 tok/s
Decode
Decode speed
854 tok/s
Prefill
Prefill speed
1.41 s
TTFT
Time to first token
17.22 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
16 days ago
Last verified
Performance
- Ollama · Q5_K_S (current)Decode 220.5 · Prefill 854 ·
- Strata · IQ3_S Decode 280 · Prefill 1358.66 ·
- Strata · Q5_K_M Decode 297.96 · Prefill 1229.8 ·
- vLLM · IQ4_XS Decode 310.8 · Prefill 1667.93 ·
- Ollama · IQ1_S Decode 316.88 · Prefill 1203.93 ·
- llama.cpp · Q8_0 Decode 162.5 · Prefill 1218.8 ·
- llama.cpp · Q5_K_M Decode 236.9 · Prefill 1776.8 ·
- llama.cpp · Q4_K_M Decode 285.3 · Prefill 2139.8 ·
Core figures
- Decode (avg)
- 220.5 tok/s
- Prefill (avg)
- 854 tok/s
- TTFT (avg)
- 1.41 s
- VRAM (avg)
- 17.22 GB
- MTP acceptance rate
- —
- TTFB
- 1.76 GB
- Power draw
- 407.2 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Llama-3.1-8B-Instruct
- Quantization
- Q5_K_S
- Framework
- Ollama
- Version
- v0.23
- Context length
- 2048 tokens
- Batch size
- 256
- GPU layers
- 33
- Flash Attention
- Off
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- NVIDIA RTX 5090
- Nominal VRAM
- 32 GB
- Measured VRAM (avg)
- 17.22 GB
- OS
- Windows 11 23H2
- Driver
- 570.86
- CUDA
- 12.4
- Power draw
- 407.2 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.23 · Windows 11 23H2 · CUDA 12.4 · 2048 ctx
220.5 tok/s
Decode
854 tok/s
Prefill
1.41 s
TTFT
17.22 GB
VRAM
—
MTP
407.2 W
Power draw
L2 batch seed 2026-10-19: Ollama on hw_id=13 running model_id=68 at Q5_K_S